CGAL users discussion list

["Sebastien Loriot (GeometryFactory)" <>]

This will be fixed as soon as https://github.com/CGAL/cgal/pull/648
will be ready.

Sebastien.

On 05/11/2016 09:29 AM, nsoonhui wrote:
>  From my code, it doesn't seem so. But I have not have the time to reproduce a
> sample for this.
>
> A comment
> <https://sourceforge.net/p/flightgear/terragear/ci/86aa9261ca6b38bf14257ae45fc7c6c1120d17d2/tree/src/Lib/terragear/mesh/tg_mesh_arrangement_cleaning.cxx?diff=b740168613e5fbfa8fe7565429d06ceba30a54af>
> on the internet seems to agree with me:
>
> +    // snap rounding notes:
> +    // 1) doesn't appear to be threadsafe
> +    // 2) no way to define the origin of snapping.  so if a point is at
> 0,0, and pixel size is 1, new point will be at 0.5, 0.5.
> +    //    We don't want this, so we translate the entire dataset back by
> 1/2 pixel size so 0,0 is still 0,0
> +
> +    lock->lock();
> +    CGAL::snap_rounding_2<srTraits, srSegmentList::const_iterator,
> srPolylineList>
> +    (srInput.begin(), srInput.end(), srOutput, 0.0000002, true, false, 5);
> +    lock->unlock();
>
> Is it possible that you make this operation thread safe?
>
>
>
> --
> View this message in context: 
> http://cgal-discuss.949826.n4.nabble.com/Is-snap-rounding-2-operation-thread-safe-tp4661900.html
> Sent from the cgal-discuss mailing list archive at Nabble.com.